Correctness verification on GSM-Symbolic
0.435LBBeaver
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| BeaverModel=Llama3.3-70B2025.12 | 0.435 | 0.454 | 0.019 | 33.33 | |
| RSModel=Llama3.3-70B2025.12 | 0.43 | 0.552 | 0.122 | 59.63 | |
| BeaverModel=Qwen3-30B-A3B2025.12 | 0.404 | 0.426 | 0.022 | 38.58 | |
| BeaverModel=Qwen2.5-14B2025.12 | 0.395 | 0.439 | 0.044 | 51.54 | |
| RSModel=Qwen3-30B-A3B2025.12 | 0.384 | 0.541 | 0.157 | 72.91 | |
| RSModel=Qwen2.5-14B2025.12 | 0.356 | 0.704 | 0.348 | 85.39 | |
| BeaverModel=Qwen3-4B2025.12 | 0.343 | 0.356 | 0.013 | 24.95 | |
| RSModel=Qwen3-4B2025.12 | 0.341 | 0.433 | 0.092 | 49.02 |